The Hsp70/J-protein machinery of the African trypanosome, Trypanosoma brucei

Abstract

The etiological agent of the neglected tropical disease African trypanosomiasis, Trypanosoma brucei, possesses an expanded and diverse repertoire of heat shock proteins, which have been implicated in cytoprotection, differentiation, as well as progression and transmission of the disease. Hsp70 plays a crucial role in proteostasis, and inhibition of its interactions with co-chaperones is emerging as a potential therapeutic target for numerous diseases. In light of genome annotations and the release of the genome sequence of the human infective subspecies, an updated and current in silico overview of the Hsp70/J-protein machinery in both T. brucei brucei and T. brucei gambiense was conducted. Functional, structural, and evolutionary analyses of the T. brucei Hsp70 and J-protein families were performed. The Hsp70 and J-proteins from humans and selected kinetoplastid parasites were used to assist in identifying proteins from T. brucei, as well as the prediction of potential Hsp70–J-protein partnerships. The Hsp70 and J-proteins were mined from numerous genome-wide proteomics studies, which included different lifecycle stages and subcellular localisations. In this study, 12 putative Hsp70 proteins and 67 putative J-proteins were identified to be encoded on the genomes of both T. brucei subspecies. Interestingly there are 6 type III J-proteins that possess tetratricopeptide repeat-containing (TPR) motifs. Overall, it is envisioned that the results of this study will provide a future context for studying the biology of the African trypanosome and evaluating Hsp70 and J-protein interactions as potential drug targets.

Electronic supplementary material

The online version of this article (10.1007/s12192-018-0950-x) contains supplementary material, which is available to authorized users.

Introduction

African trypanosomiasis is a tropical disease endemic to 37 countries in sub-Saharan Africa, and those living in rural areas that depend of farming, fishing, and hunting are most at risk (WHO 2017). As the number of new cases of human African trypanosomiasis (HAT) has decreased substantially over the decades, this neglected tropical disease is targeted for eradication by 2020 (WHO 2017). The etiological agent of this disease is an extracellular blood- and tissue-borne protozoan parasite Trypanosoma brucei. T. brucei is comprised of three subspecies: Trypanosoma brucei brucei (T. b. brucei), Trypanosoma brucei gambiense (T. b. gambiense), and Trypanosoma brucei rhodesiense (T. b. rhodesiense). T. b. brucei is responsible for animal African trypanosomiasis (AAT), also known as Nagana. T. b. gambiense causes HAT and gives rise to a chronic infection and represents over 90% of reported cases in Central and Western Africa (Simarro et al. 2008), whilst T. b. rhodesiense is mainly a zoonotic disease responsible for less than 10% of reported cases of HAT in Eastern and Southern Africa and causes an acute infection, which is rapidly fatal if untreated (Brun et al. 2010). Tools for controlling the parasitic disease are limited, due to the inability to develop a vaccine, toxicity of existing drugs, and the development of parasitic resistance (Barrett and Croft 2012). The completion of the genome sequence for T. b. brucei in 2005 and the subsequent completion of other kinetoplastid genomes have facilitated transcriptome and proteome analyses. This has led to a renewed interest in the discovery of new targets and approaches for anti-trypanosomatid drugs (Wenzler et al. 2016). Amongst the potential new drug targets, heat shock proteins represent an interesting group already validated in other disease areas (Shrestha et al. 2016).

Heat shock protein 70 (Hsp70) is a prominent protein chaperone family involved in a plethora of essential cellular functions, which include but are not limited to promoting the correct protein folding of newly synthesized polypeptides and mediating protein translocation, quality control and degradation (Bukau and Horwich 1998; Agarraberes and Dice 2001; Mayer and Bukau 2005). The functional cycle of the Hsp70 chaperone system is nucleotide-dependent and regulated by co-chaperones, such as the J-protein family and nucleotide exchange factors (NEFs). Common to all J-proteins is the possession of a conserved ~ 70-amino acid region known as the J-domain (Cheetham and Caplan 1998; Craig et al. 2006; Kampinga and Craig 2010). Apart from the J-domain, members of this co-chaperone family contain a wide variety of domains, which have been used as the basis for classification of members into four classes (I–IV) (Cheetham and Caplan 1998; Botha et al. 2007). Nucleotide exchange factors (NEFs) are a group of co-chaperones that facilitate the release of the bound substrate by accelerating ADP release that essentially primes Hsp70 for the start of the next cycle (Brehmer et al. 2001; Dragovic et al. 2006). Hsp110s are one of the major eukaryotic HSPs and are divergent members of the Hsp70 family (Easton et al. 2000). Hsp110s have been shown to be potent NEFs for Hsp70 (Dragovic et al. 2006; Raviol et al. 2006), though some Hsp110 homologues have be shown to be able to bind substrate and prevent aggregation by functioning as “holdases” (Polier et al. 2008). Thus, Hsp110s possess dual roles, as chaperones and as co-chaperones of Hsp70.

The lifecycle of T. brucei is complex as these parasites must transition between two strikingly different hosts, a cold-blooded arthropod vector and a warm-blooded mammalian host. Thus, the infectious cycle results in sudden changes in growth conditions, and exposure of the parasites to a wide variety of environmental stresses (Jones et al. 2008). The Hsp70/J-protein chaperone machinery is an integral component of the heat shock response and has been found to be conserved across organisms (Boorstein et al. 1994). However, the evolution of the Hsp70 protein family has been shown to be dynamic and highly adapted to species-specific constraints (Drini et al. 2016). This has been documented by substantial variation in Hsp70 gene copy number (Daugaard et al. 2007), phylogenetically distinct subfamilies, and the evolution of atypical protein members (Hughes 1993; Boorstein et al. 1994; Gupta and Singh 1994; Kampinga and Craig 2010; Kominek et al. 2013).

A post-genomic analysis of the molecular chaperone complements in the Tritryps, T. brucei, T. cruzi, and Leishmania major (L. major), revealed an unprecedented expansion in J-protein, Hsp70 and Hsp60 complements, indicating that these protein families may play a critical role in kinetoplastid biology (Folgueira and Requena 2007). A review of the Hsp70 superfamily in the Tritryps by Louw et al. (2010a) revealed that the protein family possessed atypical Hsp70 members and features. Subsequent reviews and updated in silico analyses of Hsp70/J-protein machinery in the annotated genome sequences of intracellular kinetoplastid parasites has been conducted (Shonhai et al. 2011; Urmenyi et al. 2014; Requena et al. 2015). However, this had not been the case for the extracellular parasite, T. brucei.

Proteomic studies have compared protein expression between lifecycle stages (Gunasekera et al. 2012; Urbaniak et al. 2012; Butter et al. 2013), including the phosphoproteome (Nett et al. 2009; Urbaniak et al. 2013). The proteome of the mitochondria is available (Panigrahi et al. 2009), including the importome (Peikert et al. 2017), respiratome (Acestor et al. 2011), and mitochondrial membranes (Acestor et al. 2009). The nuclear (Goos et al. 2017), nuclear pore (DeGrasse et al. 2008) and glycosome proteomes (Colasante et al. 2006; Güther et al. 2014) have been analysed. Proteomic data is also available for the flagellum (Broadhead et al. 2006; Subota et al. 2014) and cell surface (Shimogawa et al. 2015). Considering the medical importance of these parasites, as well as the large amount of proteomic data available, an updated investigation is timely and appropriate for Hsp70 and J-proteins in trypanosomes. This study aimed to provide an updated overview of the Hsp70/J-protein chaperone machinery in T. brucei, with respect to both T. b. brucei and T. b. gambiense. The availability of the T. b. gambiense genome sequence enabled the determination and comparative analysis of the Hsp70/J-protein chaperone machinery. Other kinetoplastids included in this study were the non-parasitic Bodo saltans (Deschamps et al. 2011) and the insect-infecting Crithidia fasciculata (reviewed in Wallace 1966). The Hsp70 and J-protein families from humans and selected kinetoplastid parasites were used to assist in identifying all T. brucei Hsp70 and J-protein members, potential Hsp70–J-protein partnerships, and the inference of the cellular function of individual members and potential partnerships. African trypanosomiasis is a tropical disease that afflicts both humans and livestock. Overall, it is envisioned that the results of this study will provide a future context for studying the biology of the African trypanosome.

Materials and methods

Database mining, sequence analyses, and the determination of kinetoplastid and human orthologues

A BLASTP search using the Hsp70 proteins from T. b. brucei obtained from previous in silico studies (Folgueira and Requena 2007; Louw et al. 2010a) and human HSPA1A as queries on the TriTrypDB (version 35) database (http://tritrypdb.org/tritrypdb/; Aslett et al. 2010) was conducted in order to determine the Hsp70 superfamily encoded on the T. b. gambiense genome, as well as identify new T. b. brucei Hsp70 protein members. The e-value was set at an intermediately stringent level of e-10 for collecting as many potential Hsp70-related sequences for further analysis. Additionally, a keyword search was performed to scan the genome of T. b. gambiense for Hsp70 genes on the TriTrypDB database using the terms: “Hsp70”, “heat shock protein”, and “molecular chaperone”. The retrieved amino acid sequences from the various keyword searches were then screened for the Hsp70 domain using SMART 7 (Simple Modular Architecture Research Tool; http://smart.embl-heidelberg.de/; Letunic et al. 2012) and PROSITE (http://prosite.expasy.org/; Sigrist et al. 2009).

A search in the annotated T. b. brucei and T. b. gambiense genome sequences on the TriTrypDB database for proteins containing the J-domain in their amino acid sequence was conducted using the J-domain (1-77aa) from Escherichia coli (E. coli) DnaJ (EcDnaJ; NP_308042.1) as a query in a pBLAST search. The common denominator for all J-proteins is the possession of a J-domain (Cheetham and Caplan 1998), and all J-proteins are divided into the four-type classes based on their structural homology to E. coli DnaJ (Cheetham and Caplan 1998; Botha et al. 2007). The keyword search using the terms “Hsp40”, “DnaJ”, “Heat shock protein”, and “molecular chaperone” were also conducted to scan the genome of T. b. brucei and T. b. gambiense for J-protein genes on the TriTrypDB database. The retrieved amino acid sequences from the various keyword searches were screened using SMART 7 (http://smart.embl-heidelberg.de/; Letunic et al. 2012) and PROSITE (http://prosite.expasy.org/; Sigrist et al. 2009) for the presence of a J-domain.

For the identification of human and selected kinetoplastid orthologues, reciprocal BLASTP was conducted. In the first query, the putative amino acid sequences of the Hsp70 and J-proteins from both T. brucei subspecies were used as queries in a BLASTP search on the National Centre for Biotechnology Information (NCBI) website (www.ncbi.nlm.nih.gov), using the default parameters. The amino acid sequences of the putative orthologues were then used as second queries in BLASTP searches using default parameters on the TriTrypDB database. If the most similar orthologue in the T. brucei subspecies was exactly the Hsp70 or J-protein sequence used as the first query, the sequence of the second query was selected as an orthologue.

Phylogenetic analyses

Phylogenetic trees were constructed to analyse the phylogenetic relationship of the Hsp70 and J-protein complements in both T. brucei subspecies. The type III J-protein subfamily was omitted from the phylogenetic analysis, as the subfamily is diverse with regard to amino acid composition and protein length, with the only common feature being the J-domain. The full-length amino acid sequences for the Hsp70 superfamily and the selected J-protein subfamilies in the selected kinetoplastid parasites were obtained from the TriTrypDB database (Aslett et al. 2010), and the human and C. fasciculata protein sequences were obtained from the National Centre for Biotechnology Information (NCBI) website (www.ncbi.nlm.nih.gov). Partial amino sequences were omitted from the analysis. Accession numbers for the Hsp70, Hsp110, and J-protein sequences used in this study are provided in Tables 1 and 2 and in the supplementary data, Table S1 and S2. Multiple sequence alignments were performed using the in-built ClustalW program (Larkin et al. 2007) with default parameters in MEGA 7 (Kumar et al. 2016), and are provided in the supplementary data, Fig. S1 and S4. Maximum-likelihood (ML) was utilized to find the best model of evolution and was selected by the Bayesian Information Criterion (BIC) implemented in MEGA 7. The amino acid–based Hsp70 and J-protein ML phylogeny was reconstructed using the JTT (Jones–Taylor–Thornton) model matrix (Jones et al. 1992) with gamma distribution shape parameter (G). Maximum likelihood phylogenetic trees were constructed using MEGA 7.0 (Le and Gascuel 2008). The accuracy of the reconstructed trees was assessed using a bootstrap test using 1000 replicates with a pairwise gap deletion mode. The phylogenetic trees for Hsp70/HSPA, Hsp110/HSPH, and J-proteins were unrooted.

Table 1.

The Hsp70/HSPA and Hsp110/HSPH proteins from Trypanosoma brucei with their putative T. cruzi, L. major, C. fasciculata, B. saltans, and H. sapiens orthologues

T. brucei		T. cruzic	L. major	C. fasciculata	B. saltans	H. sapiens
Namea	Gene IDb	Gene IDb	Gene IDb	Gene IDb	Gene IDb	Gene IDb	Localisationd	Reference
A: Hsp70/HSPA
Hsp70	Tb927.11.11330 Tbg972.11.12660	TcCLB.511211.170 Tc_MARK_1461	LmjF.28.2770	CFAC1_300044700 CFAC1_300044800	ACI15927	HSPA1A HSPA1B	CYT, NUC, GLYCO, FLAGELLAR, CELL SURFACE (BSF, PF)	Güther et al. 2014
								Subota et al. 2014
								Shimogawa et al. 2015
Hsp70.4	Tb927.7.710 Tbg972.7.640	TcCLB.503721.39 TcCLB.511257.10 TCDM_01248 Tc_MARK_1012	LmjF.26.1240	CFAC1_290020800	CUG93112	HSPA2?	CYT, NUC, CELL SURFACE (PF)	Shimogawa et al. 2015
Hsp70.c	Tb927.11.11290 Tbg972.11.12620	TcCLB.511211.220 TCDM_07862 Tc_MARK_1466	LmjF.28.2820	CFAC1_300045200	CUG86635	HSPA6?	CYT, NUC	Goos et al. 2017
GRP78A GRP78B	Tb927.11.7460 Tb927.11.7510 Tbg972.11.8650 Tbg972.11.8690	TcCLB.506585.40 TCDM_08367 Tc_MARK_6525 Tc_MARK_6614	LmjF.28.1200	CFAC1_300021300	CUE68699 CUG90530	HSPA5	ER	Bangs et al. 1993
MtHsp70A MtHsp70B MtHsp70C	Tb927.6.3740 Tb927.6.3750 Tb927.6.3800 Tbg972.6.3510 Tbg972.6.3520 Tbg972.6.3580	TcCLB.507029.30 Tc_MARK_1997 Tc_MARK_2001 Tc_MARK_2002	LmjF.30.2460 LmjF.30.2470 LmjF.30.2480 LmjF.30.2490 LmjF.30.2550	CFAC1_260048700 CFAC1_260048800 CFAC1_260049400	ACI15928	HSPA9	MITO, CELL SURFACE (BSF, PF)	Panigrahi et al. 2009
								Niemann et al. 2013
								Shimogawa et al. 2015
B: Hsp110/HSPH
Hsp110	Tb927.10.12710 Tbg972.10.15330	TcCLB.507831.60 TCDM_05266 Tc_MARK_4982	LmjF.18.1370	CFAC1_140023200	CUG91811	HSPH1	CYT, CELL SURFACE (PF)	Shimogawa et al. 2015
Grp170	Tb927.9.9860 Tbg972.9.5670	TcCLB.506885.440 TcCLB.508457.20 TCDM_ 08776 Tc_MARK_4314	LmjF.35.4710	CFAC1_300098600	CUG89677	HSPH4	ER	Field et al. 2010
Hsp70.a	Tb927.9.4500 Tbg972.9.2380	TcCLB.511585.70 TcCLB.510155.70 TCDM_06223 Tc_MARK_2135	LmjF.01.0640	CFAC1_050012600	CUF98600	–	ER, MITO	Field et al. 2010
								Acestor et al. 2009;
								Panagrahi et al. 2009
								Peikert et al. 2017
								Niemann et al. 2013
Hsp70.b	Tb927.7.1030 Tbg972.7.980	TcCLB.503899.10 TCDM_01297 Tc_MARK_1045	LmjF.26.0900	CFAC1_290016600	–	–	MITO	Peikert et al. 2017
								Niemann et al. 2013

CYT cytosol, MITO- mitochondrion, NUC- nucleus, ER- endoplasmic reticulum, GYLCO- glycosomes, BSF- bloodstream form, PF- procyclic form

^aThe nomenclature for the Hsp70/HSPA and Hsp110/HSPH proteins from T. b. brucei and T. b. gambiense were derived according to Folgueira and Requena (2007)

^bThe gene IDs for the members of the T. b. brucei, T. b. gambiense, T. cruzi, C. fasciculata, and L. major Hsp70/HSPA and Hsp110/HSPH protein family were retrieved from the TriTrypDB database (http://tritrypdb.org/tritrypdb/; Aslett et al. 2010). The Gene IDs for the members of the B. saltans and H. sapiens Hsp70/HSPA and Hsp110/HSPH protein family were retrieved from NCBI (https://www.ncbi.nlm.nih.gov/)

^cThe gene IDs for the orthologues, identified by reciprocal BLASTP analysis, of three strains of T. cruzi are listed. T. cruzi CL Brener Esmeraldo-like (TcCLB), T. cruzi Dm28c (TCDM), and T. cruzi marinkelli strain B7 (Tc_MARK)

^dSubcellular localizations for the T. brucei J-proteins were either predicted using the online prediction servers and/or determined using various proteomic datasets listed in the materials and methods

Table 2.

The J-proteins from Trypanosoma brucei with their putative T. cruzi, L. major, C. fasciculata, B. saltans, and H. sapiens orthologues

T. bruceia			T. cruzicd	L. majorc	C. fasciculatac	B. saltansc	H. sapiensc
Type	Gene IDb	Name	Gene IDb	Gene IDb	Gene IDb	Gene IDb	Name	Localisatione	Functionf	Reference
I	Tb927.2.5160 Tbg972.2.3190	J2	TcCLB.511627.110 Tc_MARK_9726 TCDM_08977	LmjF.27.2400	CFAC1_210029700	CUG85902	DnaJA1	CYT	Protein aggregation and refolding	Ludewig et al. 2015
										Terada et al. 2005
										Ahrendt and Braun 2010
										Burger et al. 2014
	Tb927.10.2290 Tbg972.10.2780	J3	TcCLB.511367.138 Tc_MARK_5591 TCDM_04677	LmjF.21.0490	CFAC1_260011100	CUE68025	–	CYT	Protein aggregation and refolding	Ashburner et al. 2000
	Tb927.9.8410 Tbg972.9.4590	J27	TcCLB.510243.30 Tc_MARK_5901 TCDM_05671	LmjF.04.0940	CFAC1_030019200	CUE72942	–	MITO	Protein aggregation and folding mtDNA maintenance	Acestor et al. 2009, 2011
										Panigrahi et al. 2009
										Niemann et al. 2013
	Tb927.11.16740 Tbg972.11.18880	J45	TcCLB.511531.9 Tc_MARK_2746 TCDM_02688	LmjF.32.3300	CFAC1_300054500	CUG88240	–	ER MITO	ERAD	Schlenstedt et al. 1995
										Panigrahi et al. 2009
										Niemann et al. 2013
	Tb927.3.1430 Tbg972.3.1180	J46	TcCLB.509437.40 Tc_MARK_5358 TCDM_10337	LmjF.25.1100	CFAC1_220020000	CUG87582	DnaJB11	ER	Protein folding mRNA editing	Jin et al. 2009
	Tb927.9.12730 Tbg972.9.7790	J50	TcCLB.510743.100 Tc_MARK_4474 TCDM_00202	LmjF.35.2980	CFAC1_300079900	CUG89607	DnaJA3	MITO	Protein aggregation and folding mtDNA maintenance	Týč et al. 2015
										Acestor et al. 2009
										Niemann et al. 2013
II	Tb927.10.8540 Tbg972.10.10330	J6	TcCLB.506355.50 Tc_MARK_3436 TCDM_11062	LmjF.36.6270	CFAC1_280070900	CUG94252	DnaJB1	CYT	Protein (re)folding	Freeman and Morimoto 1996
	Tb927.11.15130 Tbg972.11.17000	J7	TcCLB.509157.80 Tc_MARK_2851 TCDM_10924	LmjF.32.1940	CFAC1_190031300	CUG94211	–	MITO FLAGELLAR	?	Acestor et al. 2009
										Panigrahi et al. 2009
										Niemann et al. 2013
										Subota et al. 2014
III	Tb927.11.16980 Tbg972.11.19710	J1	TcCLB.511537.50 Tc_MARK_2772 TCDM_02646	LmjF.32.3030	CFAC1_300057600	CUE71254	–	CYT, NUC	Protein (re)folding	Louw et al. 2010b
	Tb927.10.5660 Tbg972.10.6880	J5	TcCLB.504163.100 Tc_MARK_8575 TCDM_01101	LmjF.36.1330	CFAC1_250022200	CUG21319	–	MITO	Protein folding Protein import-MITO	Panigrahi et al. 2009
										Niemann et al. 2013
	Tb927.9.8410 Tbg972.9.4590	J8	TcCLB.510243.30 Tc_MARK_5901 TCDM_05671	LmjF.04.0940	CFAC1_210013600	CUE72942	DnaJC15?	MITO, GLYCO	Protein aggregation and folding mtDNA maintenance	Hatle et al. 2013
										Acestor et al. 2009
										Niemann et al. 2013
										Güther et al. 2014
	Tb927.6.3850 Tbg972.6.3630	J9	TcCLB.511517.44 Tc_MARK_2007 TCDM_02176	–	CFAC1_140009800	ACI15980	–	MITO	?	Peikert et al. 2017
	Tb927.7.6660 Tbg972.7.760	J10	TcCLB.510055.140 Tc_MARK_7736 TCDM_09211	LmjF.17.0460	CFAC1_090011100	CUG14142	–	MITO		Acestor et al. 2009
										Peikert et al. 2017
	Tb927.9.8160 Tbg972.9.4440	J11	TcCLB.510241.70 Tc_MARK_5912 TCDM_05687	LmjF.04.0780	CFAC1_030017200	CUG91719	–	MITO		Acestor et al. 2009
										Niemann et al. 2013
	Tb927.4.2970 Tbg972. 4.2970	J12	TcCLB.506435.50 Tc_MARK_2622 TCDM_03557	–	CFAC1_230048800	CUG94092	–	MITO	?	–
	Tb927.10.12530 Tbg972.10.12530	J13	TcCLB.509809.20 Tc_MARK_4997	LmjF.18.1490	CFAC1_140024700	CUG91826	DnaJC24?	CYT, NUC	Dipthamide synthesis	Liu et al. 2004
	Tb927.5.2880 Tbg972.5.4050	J14	TcCLB.507063.180	LmjF.08.0990	CFAC1_200014300	CUG33795	–	CYT, NUC	?	–
	Tb927.10.14730 Tbg972.10.17890	J15	TcCLB.503833.20 Tc_MARK_4789 TCDM_00497	LmjF.19.0080	CFAC1_170007200	CUG02245	–	CYT	?	Nett et al. 2009
	Tb927.1.1960 Tbg972.1.1130	J16	TcCLB.506473.20 Tc_MARK_6047 TCDM_00922	LmjF.20.1130	CFAC1_180019100	CUI12704	DnaJC2	CYT	Translation	Otto et al. 2005
										Nett et al. 2009
	Tb927.1.4300 Tbg972.1.2850	J17	TcCLB.506529.260 Tc_MARK_5675 TCDM_05005	LmjF.12.1110	CFAC1_010014800	CUG89438	–	CYT, NUC	?	–
	Tb927.11.1010 Tbg972.11.1070	J18	TcCLB.506925.470 Tc_MARK_6358 TCDM_01954	LmjF.27.0410	CFAC1_230048800	CUG94092	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Acestor et al. 2009
										Peikert et al. 2017
	Tb927.4.650 Tbg972.4.400	J19	TcCLB.507053.120 Tc_MARK_2469	LmjF.34.4080	CFAC1_290072000	CUI15604	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Panigrahi et al. 2009
										Peikert et al. 2017
	Tb927.10.5040 Tbg972.10.6100	J20	TcCLB.510293.50 Tc_MARK_8503 TCDM_01182	LmjF.36.0610	CFAC1_250013400	CUG62628	–	MEM	Protein import	Ashburner et al. 2000
	Tb927.7.540 Tbg972.7.420	J21	TcCLB.506287.90 Tc_MARK_965 TCDM_8503	LmjF.26.1410	CFAC1_290023200	CUI15434	–	MITO, GLYCO	Protein import	Panigrahi et al. 2009
										Güther et al. 2014
										Peikert et al. 2017
										Niemann et al. 2013
	Tb927.10.6610 Tbg972.10.8080	J22	TcCLB.510187.330 Tc_MARK_3742 TCDM_06560	LmjF.36.2110	CFAC1_250032000	CUG93498	–	MITO, GLYCO	?	Niemann et al. 2013
										Zíková et al. 2017
										Güther et al. 2014
	Tb927.10.13830 Tbg972.10.17000	J23	TcCLB.507993.30 Tc_MARK_4871	LmjF.18.0330	CFAC1_140009800	ACI15980	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Acestor et al. 2009
										Niemann et al. 2013
	Tb927.6.3120 Tbg972.6.2910	J24	TcCLB.511751.60 Tc_MARK_1932	LmjF.30.1790	CFAC1_260039300	CUG86810	–	CYT, NUC	?	–
	Tb927.7.680 Tbg972.7.590	J25	TcCLB.506289.74 Tc_MARK_979 TCDM_07562	LmjF.26.1270	CFAC1_290021600	CUG93121	–	MITO	?	Panigrahi et al. 2009
										Niemann et al. 2013
	Tb927.7.6200 Tbg972.7.7170	J26	TcCLB.506513.30 Tc_MARK_7763	LmjF.17.0040	CFAC1_090006000	CUG74938	–	MITO, GLYCO	Protein import-MITO	Ashburner et al. 2000
										Acestor et al. 2009
										Güther et al. 2014
										Niemann et al. 2013
	Tb927.7.740 Tbg972.7.670	J28	TcCLB.506135.40 TCDM_01252	LmjF.26.1200	CFAC1_290020200	CUG93098	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Panigrahi et al. 2009
										Peikert et al. 2017
	Tb927.11.5710 Tbg972.11.6430	J29	TcCLB.506203.50 Tc_MARK_311 TCDM_00777	LmjF.24.1080	CFAC1_210020200	CUF82364	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Panigrahi et al. 2009
	Tb927.8.1010 Tbg972.8.590	J30	TcCLB.508569.120 Tc_MARK_8237 TCDM_01742	LmjF.07.0780	CFAC1_080017400	CUI14395	–	MITO	Protein import-MITO	Ashburner et al. 2000
										Panigrahi et al. 2009
										Peikert et al. 2017
	Tb927.3.2290 Tbg972.3.2270	J32	TcCLB.508479.280 Tc_MARK_5111 TCDM_08321	LmjF.25.2190	CFAC1_230010600	CUG43629	–	ER	?	Field et al. 2010
	Tb927.10.9840 Tbg972.10.12010	J33	TcCLB.504147.60 Tc_MARK_2256 TCDM_02891	LmjF.36.4470	CFAC1_280050000	CUG72683	–	CYT	?	–
	Tb927.9.10010 Tbg972.9.5780	J34	TcCLB.506887.90 Tc_MARK_4329 TCDM_00009	LmjF.35.4630	CFAC1_300097800	CUG92668	–	ER, MITO		Goldshmidt et al. 2008
										Acestor et al. 2009
										Panigrahi et al. 2009
										Peikert et al. 2017
										Niemann et al. 2013
	Tb927.7.4590 Tbg972.7.5200	J35	TcCLB.506605.150 Tc_MARK_7879 TCDM_05890	LmjF.14.0110	CFAC1_110006300	CUF84499	–	MITO	?	Panigrahi et al. 2009
	Tb927.3.1760 Tbg972.3.1610	J36	TcCLB.510091.50 Tc_MARK_5057 TCDM_03945	LmjF.25.1690	CFAC1_230016200	–	DnaJC20	MITO	FeS cluster biogenesis	Mokranjac et al. 2003
										Uhrigshardt et al. 2010
										Acestor et al. 2009
	Tb927.10.12640 Tbg972.10.15230	J37	TcCLB.503455.10 Tc_MARK_4988 Tc_MARK_4989 TCDM_05256	LmjF.18.1430	–	CUG91818	–	CYT, NUC	?	–
	Tb927.6.3730 Tbg972.6.3500	J38	TcCLB.506941.270	LmjF.30.2450	CFAC1_140023800	CUF23573	–	MITO	Protein import-MITO	Ashburner et al. 2000
Panigrahi et al. 2009
Peikert et al. 2017
Tb927.9.14180 Tbg972.9.8890	J39	TcCLB.510759.134 Tc_MARK_4558 TCDM_03405	–	CFAC1_250032000	–	–	MITO	?	Niemann et al. 2013
Tb927.8.4470 Tbg972.8.4230	J40	TcCLB.508989.60	LmjF.10.1050	CFAC1_040019100	CUG90885	–	MITO	?	Peikert et al. 2017
Tb927.4.3980 Tbg972.4.4110	J41	–	–	CFAC1_270012300	CUG89969	–	CYT, GOLGI	Intracellular trafficking	Ashburner et al. 2000
Tb927.10.12380 Tbg972.10.14920	J42	TcCLB.507625.110 Tc_MARK_5009 TCDM_09879	LmjF.18.1650	CFAC1_140028800	–	–	CYT	Protein folding	Ashburner et al. 2000
Tb927.9.10950 Tbg972.9.6470	J43	TcCLB.508461.240 Tc_MARK_4379 TCDM_00080	LmjF.35.4040	CFAC1_300054500	–	–	CYT	?	–
Tb927.8.7010 Tbg972.8.7230	J44	TcCLB.509911.100 Tc_MARK_1523	LmjF.31.3100	CFAC1_270068300	CUE89851	–	MITO	Protein import-MITO	Ashburner et al. 2000
									Panigrahi et al. 2009
									Peikert et al. 2017
Tb927.8.1030 Tbg972.8.610	J48	TcCLB.508569.140 Tc_MARK_8239 TCDM_01739	LmjF.31.3100	–	–	–	MITO	Protein import-MITO	Ashburner et al. 2000
									Panigrahi et al. 2009
									Peikert et al. 2017
Tb927.6.2480 Tbg972.8.2240	J49	TcCLB.509965.229 TCDM_01497	LmjF.30.1030	CFAC1_260030200	CUG89958	–	MITO	Protein import-MITO	Ashburner et al. 2000
									Panigrahi et al. 2009
									Peikert et al. 2017
Tb927.4.2220 Tbg972.4.2160	J51	TcCLB.506559.430 Tc_MARK_452 TCDM_04700	LmjF.34.2430	CFAC1_290053200	CUG93032	–	CYT	Protein folding	Ashburner et al. 2000
Tb927. 10.4900 Tbg972.10.5950	J52	TcCLB.504203.60 Tc_MARK_8493 TCDM_01195	LmjF.36.0500	CFAC1_250012000	CUG91204	DnaJC7	CYT	Protein folding quality control	Brychzy et al. 2003
Tb927.7.3630 Tbg972.7.4040	J53	TcCLB.510407.80 Tc_MARK_8808 TCDM_06728	LmjF.14.1330	CFAC1_110026000	CUG92395	DnaJC3	ER, MITO, GLYCO	ER protein synthesis	Yan et al. 2002
									Goodman et al. 2011
									Güther et al. 2014
Tb927.11.8420 Tbg972.11.9590	J55	TcCLB.506931.30	LmjF.28.1270	CFAC1_300022100	CUF97845	–	SEC	?	–
Tb927.2.3960 Tbg972.2.2180	J56	TcCLB.511109.90 Tc_MARK_9638 TCDM_14441	LmjF.33.2690	CFAC1_210037700	–	DnaJB3?	NUC	Protein folding	Ashburner et al. 2000
Tb927.11.5920 Tbg972.11.6680	J58	TcCLB.508515.120 Tc_MARK_6411 TCDM_00738	LmjF.24.1300	CFAC1_210023600	CUG89365	DnaJC14	ER, MITO, CYT	Cell surface export	Jung et al. 2016
Tb927.6.3500 Tbg972.6.3250	J59	TcCLB.506941.9 Tc_MARK_ 1973 TCDM_06339 TCDM_06340	LmjF.30.2210	CFAC1_260045500	CUF24956	DnaJC13	CYT	Endosome trafficking	Girard et al. 2005
									Girard and McPherson 2008
Tb927.11.13600 Tbg972.11.15180	J60	TcCLB.506753.160 Tc_MARK_5565 TCDM_05717	LmjF.08.0650	CFAC1_050025800	CUG87832	–	MITO	?	Panigrahi et al. 2009
Tb927.10.2460 Tbg972.10.3050	J62	TcCLB.412943.9 Tc_MARK_3379 TCDM_04012 TCDM_04013	LmjF.34.0040	CFAC1_290027800	CUF08277	–	MEM	Protein import	Ashburner et al. 2000
Tb927.11.13830 Tbg972.11.15450	J63	TcCLB.509161.110 Tc_MARK_2986 TCDM_09129	LmjF.32.0590	CFAC1_190012200	CUI15501	–	MITO		Acestor et al. 2009
Tb927.4.880 Tbg972.4.680	J65	TcCLB.508257.160 Tc_MARK_2488 TCDM_02481	LmjF.34.3870	CFAC1_290069300	CUI15574	–	MITO	Protein folding	Ashburner et al. 2000
Tb927.10.5180 Tbg972.10.6290	J67	TcCLB.510297.30 TCDM_01165	LmjF.36.0760	CFAC1_250015000	CUG94364	–	MITO	Protein folding	Ashburner et al. 2000
Tb927.11.10950 Tbg972.11.12260	J69	TcCLB.509823.10 Tc_MARK_7219 TCDM_09821	LmjF.36.4970	CFAC1_280056000	CUG89391	–	MITO	?	Panigrahi et al. 2009
Tb927. 8.8310 Tbg972. 8.8650	J70	–	–	–	–	–	GLYCO	?	–
Tb927. 9.12530 Tbg972.9.7690	J71	TcCLB.510741.190 Tc_MARK_4463 TCDM_00188	LmjF.35.3090	CFAC1_300081200	CUG89593	–	MITO	Protein import-MITO	Ashburner et al. 2000
Tb927.11.9060 Tbg972.11.10120	J72	TcCLB.507077.30 Tc_MARK_1361 TCDM_07079	–	–	–	–	CYT	?	–
Tb927.9.1560 Tbg972.9.280	J73	TcCLB.510347.50 Tc_MARK_2404 TCDM_08512	LmjF.26.2520	CFAC1_160029000	CUG90840	–	NUC		Goos et al. 2017
IV	Tb927.7.990 Tbg972.7.940	J31	TcCLB.506729.50 Tc_MARK_1042 TCDM_01293	LmjF.26.0940	CFAC1_290017000	–	–	MITO	?	Acestor et al. 2009
										Panigrahi et al. 2009
										Niemann et al. 2013
	Tb927.1.1230 Tbg972.1.570	J47	TcCLB.511423.170 Tc_MARK_6003 TCDM_01017	LmjF.20.0550	CFAC1_180012900	CUG67522	–	MITO	Protein folding	Panigrahi et al. 2009
	Tb927.8.6310 Tbg972.8.6370	J68	TcCLB.503885.70	LmjF.24.1910	CFAC1_250049500	CUG31915	DnaJC19	MITO	Protein import-MITO	Ashburner et al. 2000

CYT cytosol, MITO mitochondria, NUC nucleus, ER endoplasmic reticulum, MEM plasma membrane/membrane bound, GOLGI Golgi apparatus, GYLCO glycosomes, SEC secreted

^aThe nomenclature for the J-proteins of T. b. brucei and T. b. gambiense were derived from Folgueira and Requena (2007)

^bGene IDs for the J-proteins from T. b. brucei, T. b. gambiense, T. cruzi, L. major and C. fasciculata were obtained from TriTrypDB database (http://tritrypdb.org/tritrypdb/; Aslett et al. 2010). The gene IDs for the J-proteins from B. saltans were retrieved from NCBI (https://www.ncbi.nlm.nih.gov/)

^cOrthologues identified from T. cruzi, L. major, C. fasciculata, and Homo sapiens were determined by reciprocal BLASTP analysis

^dThe gene IDs for orthologues identified in three strains of T. cruzi are listed. T. cruzi CL Brener Esmeraldo-like (TcCLB), T. cruzi Dm28c (TCDM), and T. cruzi marinkelli strain B7 (Tc_MARK)

^eSubcellular localizations for the T. b. brucei and T. b. gambiense J-proteins were either predicted using the online prediction servers and/or determined using various proteomic datasets listed in the materials and methods

^fThe predicted cellular role and functions for each J-protein from T. brucei were implied from either Gene Ontology (Ashburner et al. 2000), or published literature on the identified functions/cellular roles of their identified orthologues

Protein properties, protein expression, domain mapping, and determination of the organelle distribution for the Hsp70 and J-protein complements

The molecular weight (Da) and isoelectric point (pI) of each gene was calculated using the compute pI/Mw tool from ExPASy (https://web.expasy.org/compute_pi/; Gasteiger et al. 2005). The protein expression level between lifecycle stages for each member of the Hsp70 and J-protein complement between lifecycle stages was retrieved from several proteomic datasets (Gunasekera et al. 2012; Urbaniak et al. 2012; Butter et al. 2013). Data on the phenotypic knockdown screen, using RNAi conducted by Alsford et al. (2011), for each member of the Hsp70 and J-protein complement was retrieved from the TrypsNetDB database (http://trypsnetdb.org/QueryPage.aspx; Gazestani et al. 2017). The protein domain mapping for the Hsp70 and J-protein complements was conducted using a combination of online programs that included TPRpred (http://toolkit.tuebingen.mpg.de/tprpred; Karpenahalli et al. 2007), SMART 7 (http://smart.embl-heidelberg.de/; Letunic et al. 2012), and PROSITE (http://prosite.expasy.org/; Sigrist et al. 2009).

Proteomic data from the mitochondrion (Panigrahi et al. 2009), mitochondrial importome (Peikert et al. 2017), respiratome (Acestor et al. 2011), mitochondrial membranes (outer, intermembrane space, inner, and matrix) (Acestor et al. 2009), nucleus (Goos et al. 2017), nuclear pore (DeGrasse et al. 2008), glycosomes (Colasante et al. 2006; Güther et al. 2014), flagellum (Broadhead et al. 2006; Subota et al. 2014), and cell surface (Shimogawa et al. 2015) were utilized for the determination of the organelle distribution for the T. brucei Hsp70 and J-protein complements. In the absence of experimental data, online prediction programs which included NucPred (http://www.sbc.su.se/~maccallr/nucpred/cgi-bin/single.cgi; Brameier et al. 2007), MitoPROT (http://ihg.gsf.de/ihg/mitoprot.html; Claros and Vincens 1996), MultiLoc (http://abi.inf.uni-tuebingen.de/Services/MultiLoc; Höglund et al. 2006), SignalP version 4.1 (http://www.cbs.dtu.dk/services/SignalP/; Petersen et al. 2011), and WoLF PSORT (http://www.genscript.com/wolf-psort.html.; Horton et al. 2007) were used.

Results and discussion

Determination of the T. b. brucei and T. b. gambiense Hsp70 superfamily

A non-human infective T. brucei subspecies, T. b. brucei, is the preferred model for trypanosome research as the T. b. brucei TREU927 strain displays the full range of known T. brucei phenotypes and possesses similar biological and genetic characteristics to the human infective subspecies, T. b. rhodesiense (Gibson 2012). The T. b. rhodesiense genome has not been sequenced, but information obtained from the T. b. brucei genome can be inferred for both subspecies (Gibson 2012). However, sequencing of the T. b. gambiense genome was conducted due to this subspecies having profoundly different biological and genetic characteristics (Jackson et al. 2010) and being the most clinically relevant subspecies, as it is the etiological agent of over 90% of HAT cases (WHO 2013). Thus, an in silico analysis of the T. brucei Hsp70 superfamily, considering annotations to the T. b. brucei genome on the TriTrypDB database, available proteomics data and the release of the genome sequence for T. b. gambiense, was conducted to provide a more current and extensive overview of the Hsp70 superfamily to previous in silico analyses.

Comparative analysis of both T. b. brucei and T. b. gambiense enabled the determination of the T. b. gambiense Hsp70 superfamily and evaluated the conservation of the T. brucei Hsp70 complement. A total of 12 putative Hsp70s were identified to be encoded on both the T. b. gambiense and T. b. brucei genomes, highlighting conservation of the Hsp70 superfamily (Table 1). This number is consistent with previous in silico studies (Folgueira and Requena 2007; Louw et al. 2010a). The Hsp70 superfamily for both T. brucei subspecies were found to comprise 8 Hsp70/HSPA proteins and 4 Hsp110/HSPH proteins (Table 1). The nomenclature for the T. b. gambiense Hsp70 superfamily was adopted from the nomenclature proposed by Folgueira and Requena (2007). However, to underscore whether discussing a protein from T. b. gambiense or T. b. brucei, the abbreviations Tbg and Tbb were used in this study respectively. Thus, TbbHsp70 refers to Hsp70 from T. b. brucei. The orthologous relationships of the Hsp70 superfamily from T. b. brucei and T. b. gambiense to the selected organisms in this study are presented in Table 1.

Three T. cruzi strains (CL Brener Esmeraldo-like, Dm28c, and marinkelli strain B7) were incorporated into this study due to the discrepancy in literature on the exact number of members for the T. cruzi Hsp70 superfamily. The Hsp70 superfamily in the T. cruzi CL Brener Esmeraldo-like strain was initially reported to comprise 28 members Folgueira and Requena 2007), though more recent studies have stated 11 members encoded on the genome (Louw et al. 2010a; Shonhai et al. 2011). This in silico study identified that the T. cruzi CL Brener Esmeraldo-like strain has 13 members, the Dm28c strain has 7 members, and the marinkelli strain B7 has 12 members. The variability amongst the three strains illustrates the need for further assessment of the Hsp70 complement in T. cruzi. The number of members for the Hsp70 superfamily in the Leishmania spp. and C. fasciculata were relatively well conserved with only variability in the gene copy numbers. In comparison to the other selected organisms in this study, the Hsp70 complement in the kinetoplastid parasites is smaller than that found in Homo sapiens (H. sapiens), which both have 17 members. Gene duplication during eukaryotic evolution satisfied the demand for Hsp70 isoforms in various intracellular compartments, tissue-specific or developmental expression patterns, and functional diversity for client specificity and/or processing in the multicellular organisms (Brocchieri et al. 2008; Kabani and Martineau 2008).

T. brucei Hsp70/HSPA subfamily

A total of 8 Hsp70/HSPA protein members were identified in this in silico study for T. b. brucei, which is consistent with previous in silico studies (Folgueira and Requena 2007; Louw et al. 2010a). Phylogenetic analysis shows that the T. brucei Hsp70/HSPA subfamily comprised 5 distinct Hsp70 groups (Hsp70, Hsp70.4, Hsp70.c, Grp78/BiP, and mtHsp70), which cluster according to protein sequence and subcellular localisation (Fig. 1). Phylogenetic analysis suggests that the T. brucei Hsp70/HSPA protein subfamily followed the same model of divergent evolution as evident amongst the other kinetoplastid parasites (Fig. 1). These five Hsp70 groups differ from each other by gene copy, protein features and domain architecture, protein expression during the lifecycle of the parasite, and the predicted or experimentally determined subcellular localisation (Table 1 and Fig. S2). These differences seem to infer that each Hsp70/HSPA protein performs a specialized cellular role(s) in the parasite. Hsp70 in both T. brucei subspecies was found to possess the canonical domain architecture of typical Hsp70s and shares high sequence identities with its orthologues in the selected organisms used in this in silico study (Fig. S2).

Fig. 1.

Phylogenetic analysis of the Hsp70 superfamily from T. brucei in relation to human and selected kinetoplastid parasites. Multiple-sequence alignment of the full-length amino acid sequences of the Hsp70/HSPA gene families in human and selected kinetoplastid parasites. The multiple sequence alignment provided in Fig. S1 was performed using the in-built ClustalW program (Larkin et al. 2007) with default parameters on the MEGA 7 software (Kumar et al. 2016). The phylogenetic tree was constructed by MEGA 7 using the Maximum-likelihood method based on the Jones–Taylor–Thornton (JTT) matrix-based model of amino acid substitution (Jones et al. 1992) with gamma distribution shape parameter (G). The alignment gaps were excluded from the analysis, and the number of amino acid sites used to construct the tree numbered 363. Bootstrap analysis was computed with 1000 replicates. Accession numbers for the T. b. brucei (Tbb), T. b. gambiense (Tbg), T. cruzi (TcCLB; CL Brener Esmeraldo), C. fasciculata (Cf), B. saltans (Bs), and L. major (Lmj) Hsp70 and Hsp110 sequences can be found in Table 1. Accession numbers for human (Hs; H. sapiens) and other kinetoplastid HSPA/Hsp70 and HSPH/Hsp110 sequences are provided in Table S1. The subcellular localisation for Hsp70s is indicated by coloured branches. Brown: cytosolic; purple: endoplasmic reticulum; and green: mitochondrion. Scale bar represents 0.2 amino acid substitutions per site (colour figure online)

Hsp70 proteins help T. brucei to adapt to changing environmental conditions, and the levels of these proteins differ during each lifecycle stage. In T. brucei, transcriptomic data often does not correlate well with protein data, and significantly larger fold changes are seen at the protein level than at the mRNA level (Urbaniak et al. 2012; Goos et al. 2017). A genome-wide comparative proteomic study between the lifecycle stages in T. brucei revealed that TbbHsp70, TbbHsp110, TbbHsp70.a, and TbbHsp70.c were downregulated in procyclic form (PF), whilst TbbmtHsp70A and TbbHsp70.4 were upregulated in PF, though with poor correlation to mRNA abundance (Urbaniak et al. 2012). In a similar study, TbbHsp70.a and TbbHsp110 were upregulated in the bloodstream from (BSF) relative to the PF, whilst TbbmtHsp70A, B, C and TbbHsp70.4 were upregulated in the PCF relative to BSF (Butter et al. 2013). The surface of T. brucei forms a vital interface with its mammalian hosts allowing it to adapt to varied environments, and cell surface proteomic analysis revealed that mtHsp70A, B, C and Hsp70 were present in both PC and BSF stages, whilst Hsp110 and Hsp70.4 were present in PF (Shimogawa et al. 2015). Protein phosphorylation plays a crucial role in the regulation of many cellular processes, and the cytosolic proteins TbbHsp70, TbbHsp70.c, and TbbHsp70.4 were determined to have at least one phosphorylation site (Nett et al. 2009; Urbaniak et al. 2013).

TbbHsp70 (Tb927.11.11330) was reported in a previous in silico study to be an unusual cytosolic Hsp70 due to the protein being shown to possess a non-canonical C-terminal RRHI motif, instead of the highly conserved C-terminal EEVD motif (Louw et al. 2010a). Following recent annotations of the T. b. brucei genome and comparison to the Hsp70 encoded on the T. b. gambiense genome (Tbg972.11.12660), the C-terminal RRHI motif has been identified as a misannotation. This is indicated to be a result of a collapse in T. b. brucei genome assembly, which caused a frameshift in the Hsp70 coding sequence, leading to changes in the C-terminal region of the protein and elimination of the EEVD motif (Droll 2013). T. b. brucei was shown to possess 5 identical copies of the TbbHsp70 gene arranged in tandem array (Glass et al. 1986), though this was collapsed into one locus following genome assembly and is constitutively co-transcribed (Lee and Van der Ploeg 1990; Huang and Van der Ploeg 1991). Duplication of the cytosolic Hsp70 gene has been shown in the other kinetoplastid parasites such as T. cruzi (Urmenyi et al. 2014), C. fasciculata (Table 1), and L. major (Requena et al. 2015; Drini et al. 2016). Amplification of HSP genes in protozoan parasites has been identified as a means of the parasites increasing chaperone activity under stressful conditions (Wiesgigl and Clos 2001).

The mRNA of Hsp70 has been shown to be regulated by a zinc finger protein, ZC3H11, where it stabilizes the mRNA after heat shock and promotes the survival of the parasite (Droll et al. 2013). HsHSPA1A/B, human orthologues of Hsp70 in both T. brucei subspecies, are major cytosolic stress-inducible Hsp70s that protect against the harmful effects of aggregates from denatured proteins during and following environmental stresses (Hartl 1996). Phenotypic knockdown, using RNAi, on the Hsp70 gene locus (Tb927.11.11330) in T. b. brucei demonstrated that it is essential to parasite survival throughout its lifecycle (Alsford et al. 2011). Based on phylogeny and orthology, it can be inferred that Hsp70 is a crucial component of the heat shock response in T. brucei, providing cytoprotection to the parasite under stressful conditions. The orthologue of Hsp70 in several Leishmania spp. has also been linked to parasite’s resistance to pentavalent antimonial treatment, as it induces Hsp70 expression which provides stress tolerance against the drug (Brochu et al. 2004; Maharjan and Madhubala 2015; Codonho et al. 2016). TbbHsp70 was detected in glycosomes with high confidence and the flagellum (Table 1) (Güther et al. 2014; Subota et al. 2014).

In this study, Hsp70.4 forms a distinct Hsp70/HSPA group found in kinetoplastid parasites, as the proteins have no mammalian orthologues (Table 1). The Hsp70.4 protein in T. brucei shares domain architecture with typical Hsp70s but possesses a divergent C-terminal EEVD motif (Fig. S2). The variation in the C-terminal EEVD motif is observed in all the kinetoplastid orthologues of Hsp70.4; DDVD in T. evansi, TDVD in T. cruzi and T. grayi, DEVD in T. vivax, TDID in B. saltans, QDVD in C. fasciculata, and EDVD in all Leishmania spp. The divergent motifs found in the kinetoplastid are proposed to be functionally equivalent to the canonical EEVD motif (Louw et al. 2010a), though the role this variation plays in the function and protein interaction with co-chaperones has not been elucidated. A previous in silico study conducted on the Hsp70 superfamily of T. cruzi noted the absence of Hsp70.4 in the T. cruzi CL Brener Esmeraldo strain genome (Louw et al. 2010a). The absence of Hsp70.4 is a result of the TcCLBHsp70.4 being on two separate loci (TcCLB.503721.39, TcCLB.511257.10) on the CL Brener Esmeraldo strain genome, both encoding for partial amino acid sequences. Additionally, a Hsp70.4 gene was found encoded on the genome sequences in the Dm28c and marinkelli strains of T. cruzi. Thus, reannotation of these loci on the T. cruzi CL Brener Esmeraldo genome is required. T. brucei Hsp70.4 is predicted to reside in the cytosol of the parasite according to its orthology and phylogeny. The Hsp70.4 orthologue in L. major has been shown through indirect immunofluorescence (IFA) staining to reside in the cytoplasm (Searle et al. 1989; Searle and Smith 1993), and to be constitutively expressed (Simpson et al. 2006), implying that the localisation and expression profile of T. brucei Hsp70.4 may be similar. TbbHsp70.4 has been shown to be non-essential as phenotypic knockdown had no detrimental effect on the survival and fitness of the parasite at any stage of its lifecycle (Alsford et al. 2011). This may indicate that the cellular functions of TbbHsp70.4 can be compensated by the other T. b. brucei cytosolic Hsp70s. However, caution should be exercised when referring to RNAi data in this study as false negatives may arise (Subramaniam et al. 2006).

Hsp70.c may represent a novel Hsp70/HSPA subfamily found only in kinetoplastids, as no clear orthologue in humans was identified (Table 1). Using various online prediction servers, the subcellular localisation was predicted to be cytosolic and nuclear, and it was found to be part of the nuclear proteome of PF T. brucei (Goos et al. 2017). Phylogenetic analysis revealed that the Hsp70.c group formed a distinct clade, as the proteins did not phylogenetically cluster with any of the other primary Hsp70/HSPA proteins (Fig. 1), which is consistent with previous phylogenetic analyses (Louw et al. 2010a; Burger et al. 2014; Requena et al. 2015). The evolutionary divergence is a result of the Hsp70 proteins possessing an atypical substrate binding domain (SBD). TbbHsp70.c (Tb927.11.11290) was shown to lack key residues that facilitate substrate recognition and were instead replaced with acidic residues (Louw et al. 2010a). The putative SBD of the Plasmodium falciparum Hsp110c was shown to be modified to handle the asparagine repeat-rich proteome of the parasite particularly during a febrile episode (Muralidharan et al. 2012). The modification of the SBD of TbbHsp70.c could be an adaptation of this particular Hsp70 to handle specific substrates in the T. b. brucei proteome during parasite differentiation. Despite these substitutions, TbbHsp70.c was still able to suppress the aggregation of the model substrates, malate dehydrogenase and rhodanese (Burger et al. 2014). Expression of TbbHsp70.c was also shown to be slightly upregulated in BSF parasites during heat shock, indicating that it could also play a potential role in parasite cytoprotection (Burger et al. 2014). Further investigation of TbbHsp70.c, and its kinetoplastid orthologues, could elucidate the cellular roles the Hsp70 fulfils in the parasites, with regard to parasite differentiation.

TbbGrp78 was the first Hsp70 isoform to be characterized from T. b. brucei, where it was shown to be a soluble luminal resident of the ER, as the C-terminal tetrapeptide MDDL maintains its subcellular localisation (Bangs et al. 1993). Both T. brucei subspecies encoded for two copies of the ER Hsp70 isoform, Grp78 (also known as BiP) (Table 1), which are 98% identical in amino acid sequence, and appear in tandem array on the genome in both subspecies. Interestingly, the duplication event of the Grp78 gene did not occur in C. fasciculata or Leishmania, as all the Leishmania spp. investigated in this study possess only one Grp78 protein (Fig. 1). It has been proposed the two Grp78 genes may be transcribed separately due to the separation of genes on chromosome XI (Louw et al. 2010a), though both Grp78A (Tb927.11.7460) and Grp78B (Tb927.11.7510) expression in T. b. brucei was shown to be upregulated at the bloodstream stage of the parasite (Bangs et al. 1993). Grp78 was further characterized by Bangs et al. (1996), where the study showed the molecular chaperone to be involved in the transport and subsequent folding of the newly synthesized variable surface glycoprotein (VSG) in the ER lumen. The upregulation and essentiality of TbbGrp78A at the bloodstream stage of the parasite may be attributed to the rapid growth of the parasites in the mammalian host and maintenance of the variable surface glycoprotein (VSG) coat of the parasite (Bangs et al. 1993).

T. brucei and several Leishmania spp. have been shown to possess a large mitochondrial Hsp70 complement, as amplification of the mtHsp70 genes is apparently a rather frequent event in kinetoplastids, with the copy numbers ranging from 2 to 5 depending on the species (Table 1). There is variability with regard to the number of mitochondrial Hsp70s in T. cruzi, as the CL Brener Esmeraldo strain was identified in this study to have one full-length gene (TcCLB.507029.30) and three partial genes encoding for mtHsp70 (TcCLB511745.10; TcCLB432677.20; TcCLB511515.40); the marinkelli strain has three genes (TcMARK_1997; TcMARK_2001; TcMARK_2002), whereas no mtHsp70 gene was found in the Dm28c strain (Table 1). The genomes of the T. cruzi marinkelli and Dm28c strains need to be further investigated to determine if the partial sequences and absence of a mtHsp70 gene respectively are sequencing errors to resolve the discrepancy in T. cruzi strains.

T. brucei possesses three mitochondrial Hsp70 homologues (mtHsp70A, mtHsp70B, mtHsp70C) (Table 1), which have been shown to appear in tandem array on the T. b. brucei chromosome VI with identical amino acid sequences (Louw et al. 2010a). TbbMtHsp70 was shown through IFA to be well distributed throughout the mitochondrion of the parasite (Klein et al. 1995) and is also an integral component of the replication and maintenance of kinetoplast DNA (kDNA) (Týč et al. 2015). The T. cruzi orthologue, TcMtHsp70, has also been implicated in mtDNA replication (Engman et al. 1989). Analysis of the T. brucei mitochondrial outer membrane proteome revealed the presence of mtHsp70A, B, and C (Niemann et al. 2013). The mammalian orthologue of the three T. brucei mitochondrial Hsp70s was identified to be HSPA9, which has been shown to facilitate the translocation and correct folding of proteins targeted for the mitochondria (Mizzen et al. 1989; Deocaris et al. 2006) (Table 1). A putative Hsp70 escort protein orthologue (TbHep1; Tb927.3.2300) was identified to be encoded on the genomes for both T. brucei subspecies, and it should be explored if TbHep1 is required to maintain solubility and functionality of the three mitochondrial Hsp70 isoforms in T. brucei.

T. brucei Hsp110/HSPH subfamily

The Hsp110/HSPH protein family in both subspecies of T. brucei was identified to comprise four members (Table 1) (Hsp110, Grp170, Hsp70.a, and Hsp70.b). All four members were shown to be considerably longer in amino acid sequence (Fig. S3), characteristic of Hsp110/HSPH protein members (Easton et al. 2000). Hsp110 is a predicted cytosolic Hsp110/HSPH protein member (Table 1; Fig. S3) that is essential throughout the lifecycle of T. b. brucei (Alsford et al. 2011). The mRNA of Hsp110 has been shown to be enriched and stabilized following heat shock in PF parasites, indicating that TbHsp110 is involved in cytoprotection and recovery following heat shock (Droll et al. 2013). Hsp110 proteins have been shown to play an important role in thermo-resistance, and the prevention of protein aggregation (Raviol et al. 2006). Hsp110 proteins are also a major component of the Hsp70 chaperone machinery, facilitating nucleotide exchange (Easton et al. 2000). Interestingly, kinetoplastid parasites only encode for one predicted cytosolic Hsp110 protein, whereas mammalian cells encode for three Hsp110 homologues (Kampinga and Craig 2010). T. brucei Hsp110 was identified in this study to be orthologous to mammalian HSPH1 (Table 1). Human HSPH1 (also known as Hsp105) has been shown to be expressed as two different isoforms, HSPH1-α and HSPH1-β (Yasuda et al. 1995). HSPH1-α is constitutively expressed but inducible to heat shock or stress, whereas HSPH1-β is strictly heat inducible (Saxena et al. 2012). HSPH-α is shown to be involved in protein biogenesis and quality control (Saxena et al. 2012). Thus, it could be suspected that Hsp110 in T. brucei forms a partnership with the predicted cytosolic Hsp70s to regulate protein biogenesis and quality control in the cytosol of the parasite. However, the cellular functions and Hsp110–Hsp70 partnerships need to be experimentally elucidated.

Grp170 and Hsp70.a are both Hsp110/HSPH protein members predicted to reside in the ER in T. brucei, as both were shown to possess N-terminal import and C-terminal ER retention signal sequences (Fig. S3). Like Grp78, mammalian Grp170 has been demonstrated to be an ER chaperone that assists in the protein folding, assembly, and transportation of secretory or transmembrane proteins (Wang et al. 2014). Grp170 in kinetoplastid parasites phylogenetically clustered with mammalian Grp170 orthologues, suggesting that the proteins may be functionally equivalent (Fig. 1). Hsp70.a is a novel ER Hsp110/HSPH protein member that is conserved in the kinetoplastid parasites (Fig. 1). TbbHsp70.a was demonstrated to be essential in parasite differentiation (Alsford et al. 2011), as the gene expression of TbbHsp70.a is upregulated during and up to 48 h post-synchronous differentiation of the parasite from the BSF to the PF life stage (Quieroz et al. 2009). Hsp70.a may be implicated in the transportation and protein folding of secretory or transmembrane proteins that could be critical for the developmental differentiation of the parasite. Suppression of the expression of TbbHsp70.a by RNAi resulted in increased accumulation of VSG in the ER and distortion of the organelle (Field et al. 2010).

In mammalian cells, the mitochondria and endoplasmic reticulum form structural and functional linkages known as mitochondria-associated ER membranes (MAMs) which are crucial to maintain cellular homeostasis (Rowland and Voeltz 2012). TbbHsp70.a, with one transmembrane domain (TMD), was also assigned to the mitochondrial membrane of PF cells with high confidence (Acestor et al. 2009). The same protein was also detected in three other mitochondrial proteomics studies (Panagrahi et al. 2009; Niemann et al. 2013; Peikert et al. 2017). Other cytosolic Hsp70s (TbbHsp70, TbbHsp70.c, and TbHsp110) were also detected in a study of the mitochondrial outer membrane proteome (Niemann et al. 2013).

Hsp70.b is a unique Hsp110/HSPH protein member, which has been shown by phylogenetic analysis to form a distinct monophyletic group (Fig. 1). Though Hsp70.b is most notably absent in B. saltans (Table 1), Hsp70.b in T. brucei is predicted to localize in the mitochondria, as the protein was shown to possess an N-terminal positively charged leader sequence (Table 1) and detected in proteomic analyses of the mitochondria (Peikert et al. 2017; Niemann et al. 2013).

The T. brucei J-protein complement

An in silico investigation identified 67 putative J-proteins encoded on the genomes of both T. brucei subspecies (Table 2). Nomenclature proposed for the T. brucei J-proteins was based on nomenclature guidelines in Folgueira and Requena (2007). For simplicity, the T. brucei J-proteins will be referred to by name, as seen in column 3 of Table 2. Phylogenetic analysis of the selected J-protein subfamilies, as illustrated in Fig. 2, shows that the J-proteins cluster based on their different classes and subcellular localisation. All identified J-protein members were further classified into the four J-protein subfamilies, I–IV, accordingly to their identified domain architecture (Table 2). The basis for classification of the J-proteins is their homology to the prokaryotic canonical J-protein, DnaJ, which is divided into an N-terminal J-domain, glycine-phenylalanine (G/F)-rich region, zinc finger-like region (ZFLR), and a C-terminal substrate-binding domain (Cheetham and Caplan 1998). Type I J-proteins possess all these canonical domains, type II J-proteins lack the ZFLR, type III J-proteins contain only the signature J-domain which can occur anywhere along the protein sequence, and type IV proteins possess a J-domain with a compromised or absent HPD motif and may also possess domain structures from the other J-protein type subfamilies (Cheetham and Caplan 1998; Botha et al. 2007). A comprehensive domain organisation of the predicted T. brucei J-proteins is illustrated in Fig. 3.

Fig. 2.

Phylogenetic analysis of selected J-protein subfamilies from T. brucei in relation to human and selected kinetoplastid parasites. Multiple-sequence alignment of the full-length amino acid sequences of the type I, II, and IV J-protein gene families in human and selected kinetoplastid parasites. The multiple-sequence alignment provided in Fig. S4 was performed using the in-built ClustalW program (Larkin et al. 2007) with default parameters on the MEGA 7 software (Kumar et al. 2016). The phylogenetic tree was constructed by MEGA 7 using the Maximum-likelihood method based on the Jones–Taylor–Thornton (JTT) matrix-based model of amino acid substitution (Jones et al. 1992) with gamma distribution shape parameter (G). The alignment gaps were excluded from the analysis, and the number of amino acid sites used to construct the tree numbered 194. Bootstrap analysis was computed with 1000 replicates. Accession numbers for the T. b. brucei (Tbb), T. b. gambiense (Tbg), T. cruzi (TcCLB; CL Brener Esmeraldo), C. fasciculata (Cf), B. saltans (Bs), and L. major (Lmj) J-protein sequences can be found in Table 2. Accession numbers for human (Hs; H. sapiens) and other kinetoplastid J-protein sequences are provided in Table S2. The subcellular localisation for J-proteins is indicated by coloured branches. Red: cytosolic; blue: endoplasmic reticulum; green: mitochondrion. Scale bar represents 0.5 amino acid substitutions per site (colour figure online)

Fig. 3.

Schematic representation of the domain architecture of the different classes of J-proteins in T. brucei. Each protein sequence for the T. brucei J-protein family is represented by an open bar with the number of amino acids indicated on either side of the protein bar. The name of the respective J-protein is indicated on the lefthand side. The various domains are highlighted by coloured blocks within the protein bar. A key is provided to give a short description of the various domains and features. The J-proteins were also categorized based on assumed client binding ability and mechanistic mode of functioning as proposed by Kampinga et al. (2009)

The total number of members for the T. b. brucei J-protein complement is larger than the previously reported 65 members by Folgueira and Requena (2007). However, 73 J-proteins were reported in T. b. brucei by Droll et al. (2013), but data mining of the available transcriptomic data revealed only 57 J-proteins indicating a discrepancy in the reported numbers for this study. A recent in silico investigation of the L. major J-protein family has also reported a larger total number of J-protein members (n = 69) encoded on the annotated genome than previously reported (n = 66) (Requena et al. 2015). However, Shonhai et al. (2011) reported fewer J-proteins (n = 61) encoded on the T. cruzi genome. This in silico study incorporated three T. cruzi strains (CL Brener Esmeraldo–like, Dm28c, and marinkelli strain B7) for in silico analysis, and investigation of the J-protein complement revealed that the T. cruzi CL Brener Esmeraldo–like strain has 66 J-protein members, Dm28c strain has 56 members, and marinkelli strain B7 has 58 members. The variability in the total J-protein member numbers highlights the need for further assessment of the T. cruzi J-protein complement. However, the J-protein complement in the kinetoplastid parasites is greatly expanded in comparison to those found in its mammalian hosts, and the significance of this expansion is yet to be elucidated.

T. brucei type I J-protein subfamily

This study identified that the type I J-protein subfamily in T. brucei comprised 6 members: J2, J3, J27, J45, J46, and J50 (Table 2). J2 was identified to be an essential type I J-protein, as knockdown via RNAi is lethal at all life stages of the parasite (Alsford et al. 2011), and it was shown to reside in the parasite cytosol (Ludewig et al. 2015). It is implicated to be an integral component of protein biogenesis, as the T. cruzi orthologue of J2 was shown to in vivo complement the yeast type I J-protein, Ydj1, and stimulate the in vitro ATPase activity of TcHsp70 (Edkins et al. 2004). The protein levels of J2 increased in response to heat stress, suggesting that the J2 protein is critical to cytoprotection in kinetoplastid parasites (Ludewig et al. 2015). The J2 orthologue in Leishmania infantum has been implicated in the differentiation process of the parasite (Tsigankov et al. 2014). J2 has demonstrated a potential partnership with Hsp70.c (Burger et al. 2014), an Hsp70 protein shown to be essential for T. brucei differentiation (Alsford et al. 2011). Thus, this Hsp70–J-protein partnership may be an integral component of parasite differentiation, and pathogenesis.

J3 is another type I J-protein predicted to localize in the cytosol of the parasite (Table 2), based on its phylogeny (Fig. 2) and domain architecture (Fig. 3). Despite the homology to J2, the two cytosolic type I J-proteins are not functionally equivalent as knockdown of J3 only resulted in loss of fitness at the PF stage of the parasite (Alsford et al. 2011), despite being shown to be expressed at all stages of the lifecycle (Aslett et al. 2010). This could suggest that the type I J-protein is required for folding of specific client proteins that are needed for PF trypanosomes. J45 and J46 are predicted to reside in the ER (Table 2), based on their phylogeny to known ER J-proteins (Fig. 2). J46 was shown to possess an N-terminal targeting sequence and similar domain architecture to its predicted human orthologue, DnaJB11 (Table 2). HsDnaJB11 is an abundant soluble ER resident type I J-protein that has been shown to co-ordinate with BiP in facilitating the folding of proteins (Jin et al. 2009; Guo and Snapp 2013).

The Hsp70/J-protein machinery was found to be indispensable for proper mitochondrial DNA maintenance and replication, as RNAi-mediated knockdown resulted in shrinkage of the highly compacted mitochondrial network, due to decreased maxicircle and minicircle copy numbers (Týč et al. 2015). J27 and J50 were predicted to reside in the mitochondrion of the parasite, based on the identification of an N-terminal targeting sequence in both J-proteins (Table 2; Fig. 3). HsDnaJA3 was identified to be the human orthologue of J50 that has been shown to co-operate with mitochondrial Hsp70s in protein translocation and folding (Iosefson et al. 2012). J50 was shown to localize in the mitochondrion of the parasite, where it was shown to form a complex with the mitochondrial Hsp70 and the nucleotide exchange factor, Mge1 (Týč et al. 2015). Analyses of the total T. brucei mitochondrial genome and the specific mitochondrial respiratory complexes in PF forms revealed the presence of a putative DnaJ protein (J27) as part of the mitochondrial respiration complex I (Panigrahi et al. 2009; Acestor et al. 2011). Proteomic analyses of three lifecycle stages showed an increase in mitochondrial protein abundance of MtHsp70A/B/C, J50, and J27 in the short stumpy and PF cells, relative to the long slender bloodstream form. A genome-wide comparative proteomic study between the lifecycle stages in T. brucei revealed that J2 and J45 were downregulated in PF form, whilst J3, J27, and J50 were upregulated in PF with an overall poor correlation to mRNA abundance (Urbaniak et al. 2012). In a similar study, J2, J45, and J46 were upregulated in the BSF relative to the PC form whilst J3, J27, and J50 were upregulated in the PCF relative to BSF (Butter et al. 2013).

T. brucei type II J-protein subfamily

Remarkably, the type II J-protein subfamily comprised only two members, J6 and J7. J7 was assigned to the mitochondrial matrix (Acestor et al. 2009) and identified in mitochondrial enriched fractions with poor signal peptide correlation (Panigrahi et al. 2009). J6 is an orthologue of Tcj6, a T. cruzi type II J-protein shown to be associated with ribosomal subunits, 80S monosomes, and smaller polysomes, and able to functionally in vivo complement a yeast mutant deficient in the orthologous gene Sis1 (Salmon et al. 2001). Tcj6 (TcCLB.506355.50) was also shown to be cytosolic, particularly concentrated around the nucleus with probable association with the ER (Salmon et al. 2001). It is speculated that J6 may perform a similar role in the cytosol of T. brucei. J7 is an essential bloodstream stage J-protein (Alsford et al. 2011). A genome-wide comparative proteomic study between the lifecycle stages in T. brucei revealed that J6 and J7 were downregulated in PF form (Urbaniak et al. 2012). In a similar study, J7 was upregulated in the BSF relative to the PC form (Butter et al. 2013).

T. brucei type III J-protein subfamily

A total of 56 members identified in the T. brucei J-protein family were found to be classified as type III J-protein members (Table 2), as these members were identified to possess a wide variety of protein domains and motifs, as illustrated in Fig. 3. The functional diversity of the type III J-protein subfamily enables the Hsp70/J-protein chaperone machinery to perform a diverse range of functions within the cell (Kaschner et al. 2015). Predicted subcellular localisation of the type III J-protein subfamily indicates that the family members localize to various organelles within the parasite (Table 2), with the majority of J proteins localized in the mitochondrion. A genome-wide comparative proteomic study between the lifecycle stages in T. brucei revealed that J20, J22, J24, J34, and J53 were downregulated in PF, whilst J1, J5, J8, J10, J11, J15, J16, J18, J21, J23, J25, J26, J28, J33, J36, J39, J51, J52, and J70 were upregulated in PF with poor correlation to mRNA abundance (Urbaniak et al. 2012). In a similar study, J1, J6, J14, J15, J20, J22, J24, J25, J34, J53, and J59 were upregulated in the BSF relative to the PC form whilst J5, J8, J10, J11, J16, J18, J21, J23, J28, J30, J32, J33, J36, J38, J39, J44, J48, J51, J52, J56, J63, J69, J70, and J73 were upregulated in the PCF relative to BSF (Butter et al. 2013). The cytosolic J15 and J16 proteins were identified as phosphoproteins with at least one phosphorylation site (Nett et al. 2009; Urbaniak et al. 2013). In a further study, a comparison of the phosphoproteins in two lifecycle stages was carried out and phosphorylation of J12, J25, J33, and J37 resulted in a 10-fold upregulation in BSF relative to PC, whilst phosphorylation of J32 resulted in a 10-fold upregulation in PC relative to BSF (Urbaniak et al. 2013). Additional J-proteins that were identified to be phosphorylated include J8, J11, J12, J14, J24, J25, J33, J34, J37, J43, J44, J51, and J59 (Urbaniak et al. 2013).

J1 was shown to be expressed in the T. b. brucei BSF stage and was unable to stimulate the ATPase activities of two different Hsp70s and did not possess independent chaperone activity, as observed for type I and II J-proteins (Louw et al. 2010b). This is not surprising as in silico analysis of the domain architecture of J1 showed the absence of a substrate-binding domain (Fig. 3). J11 was identified to be a palmitoylated protein during an analysis of palmitoylation in T. brucei (Emmer et al. 2011). J34 (also known as TbbSec63), an orthologue of ScSec63 and HsDnaJC23, is an ER membrane bound J-protein that is a component of the ER translocon, an oligomeric protein translocation pore complex that facilitates the translocation of secretory protein precursors across the ER (Engstler et al. 2007; Goldshmidt et al. 2008). RNAi-mediated knockdown of J34 was shown to be lethal (Goldshmidt et al. 2008; Alsford et al. 2011), as it affected the entry of both N-terminal ER signal peptide-containing proteins and polytopic membrane proteins (Goldshmidt et al. 2008). J34 has been implicated along with several other predicted ER chaperones to facilitate the biosynthesis and quality control of VSG proteins (Field et al. 2010). RNAi-mediated knockdown of J34 and TbbGrp78 was shown to impair protein secretion, cell viability, and presentation of variant surface glycoproteins (Field et al. 2010).

There are 6 T. brucei type III J-proteins (J42, J51, J52, J53, J65, and J67) that possess tetratricopeptide repeat-containing (TPR) motifs (Fig. 3). The tetratricopeptide repeat (TPR) is a protein–protein interaction motif that comprises a degenerate 34-amino acid sequence and has been found in many diverse proteins in all organisms (Lamb et al. 1995; D'Andrea and Regan 2003). The mammalian system is shown to possess only two TPR-containing J-proteins, DnaJC7 and DnaJC3 (Kampinga and Craig 2010). DnaJC3 (also referred to as ERdj6 or p58IPK) is a prominent type III J-protein family member in the ER, where it functions as a co-chaperone and regulator of GRP78/BiP, aiding in the refolding of misfolded proteins and thus restoring ER homeostasis (Rutkowski et al. 2007; Petrova et al. 2008). J53 was identified in this study to be the putative orthologue of DnaJC3, as it was shown to possess an N-terminal ER signal peptide (Fig. 3).

DnaJC7 (also referred to as Tpr2 or p60) is a ubiquitously expressed TPR-containing J-protein in the cytosol (Murthy et al. 1996; Ohno et al. 2014). This type III J-protein has been shown to possess two TPR domains that bind Hsp70 and Hsp90 indiscriminately, where it has been proposed to catalyze the retrograde transfer of client proteins from Hsp90 back to Hsp70 (Brychzy et al. 2003). Thus, DnaJC7 has been proposed to be a sensor of folding quality within the Hsp90 chaperoning system (Brychzy et al. 2003; Moffatt et al. 2008). Three TPR-containing J-proteins (J42, J51, and J52) have been predicted to reside in the cytosol of the parasite (Table 2). J52 was identified to be the putative orthologue of DnaJC7, but it may be speculated that the expansion of the numbers of cytosolic TPR-containing J-proteins is to offer specificity to the Hsp70/Hsp90 multichaperone heterocomplex with regard to mediating quality control of client proteins. Interestingly, only J51 has been shown to be essential to the parasite, where knockdown was shown to be lethal at the bloodstream and differentiation stages (Alsford et al. 2011). J65 and J67 are predicted to localize to the mitochondrion (Table 2). It would be worth investigating if the TPR domains of both J65 and J67 are able to act as a docking site for interaction with the mitochondrial Hsp70s, and the mitochondrial Hsp90 paralogue, TRAP-1 or HSP75.

T. brucei type IV J-protein subfamily

The T. brucei type IV J-protein subfamily comprised J31, J47, and J68, as these J-proteins were identified to possess J-domains that had abrogated HPD motifs. J68 was identified to be orthologous to HsDnaJC19 (Table 2). HsDnaJC19 is involved in the translocation of proteins into the mitochondria (Davey et al. 2006), and it can be inferred that J68 performs a similar role in the parasite. J31 was a predicted cytosolic protein, but it was assigned to the mitochondrial matrix (Acestor et al. 2009) and identified in mitochondrial enriched fractions with poor signal peptide correlation (Panigrahi et al. 2009) (Table 2). A genome-wide comparative proteomic study between the lifecycle stages in T. brucei revealed that J31 was upregulated in PF (Urbaniak et al. 2012). In a similar study, J31 and J68 were upregulated in PC form relative to the BSF (Butter et al. 2013). A comparison of the phosphoproteins in two lifecycle stages was carried out, and phosphorylation of J31 resulted in a 10-fold upregulation in BSF relative to PC (Urbaniak et al. 2013). J47 is a mitochondrial protein (Table 2) that remarkably possesses all the domains of a canonical type I J-protein. Phylogenetic analysis revealed that J47 forms a monophyletic clade with the predicted mitochondrial type I J-protein, J27 (Fig. 3). It could be assumed that J47 is a type I J-protein, with the absent HPD motif being the result of a sequencing error. However, investigation of the kinetoplastid orthologues of J47 shows that the abrogated J-domain is conserved. Investigation into the role this J-protein plays in kinetoplastid biology and its interaction with Hsp70 chaperone partners need to be elucidated.

Conclusion

This in silico study aimed to investigate the Hsp70/J-protein chaperone machinery in the T. b. brucei-annotated genome sequence, as well as to be the first to determine the Hsp70 and J-protein complements in the human infective subspecies, T. b. gambiense. These complements were comparatively analyzed in both subspecies and shown to be conserved. The T. brucei Hsp70 complement was found to comprise 12 members, with 4 belonging to the Hsp110/HSPH subfamily. This is consistent with the findings in previous in silico studies (Folgueira and Requena 2007; Louw et al. 2010a). Examination of the amino acid sequence of TbbHsp70 showed that the protein possesses a C-terminal EEVD motif, as opposed to a RRHI motif stated by Louw et al. (2010a). The misannotation is a result of a frameshift in the coding sequence after collapse of the five genes into one locus in genome assembly. Phylogenetic analysis revealed that the T. brucei Hsp70/HSPA family comprised five distinct Hsp70 groups, with multiple copies for the mitochondrial and ER Hsp70 isoforms. Hsp70.c and Hsp70.4 were both indicated to be novel cytosolic Hsp70 subgroups, as the Hsp70 proteins were only found in kinetoplastid parasites, and that the members of these Hsp70 group were found to possess atypical Hsp70 features. It is tempting to speculate that the genetic adaptation of the Hsp70 superfamily in kinetoplastid parasites is a means of coping with the environmental stresses the parasites encounter during their infectious lifecycle.

In this study, the T. brucei J-protein complement was identified to comprise 67 members. The total number of members for the T. b. brucei J-protein complement is larger than the previously reported 65 members by Folgueira and Requena (2007), with three new J-proteins (J71, J72, and J73) being identified in the annotated T. b. brucei genome sequence. Though 73 J-proteins were reported by Droll et al. (2013), there is a discrepancy to the reported numbers as data mining of the dataset revealed only 58 members. The J-protein family in T. cruzi was also identified in this study to be larger than in previously reported in silico studies (Folgueira and Requena 2007; Shonhai et al. 2011; Requena et al. 2015). However, there is discrepancy with regard to the total number of members for both Hsp70 and J-protein families in T. cruzi, as the number of members was variable in the three strains (CL Brener Esmeraldo-like, Dm28c, and marinkelli strain B7) used in this study. The T. cruzi species displays a considerable genetic and phenotypic diversity (Dvorak 1984; Tibayrenc 1998), a result of a predominantly clonal mode of evolution through large time spans (Tibayrenc et al. 1986; Tibayrenc and Ayala 2002). The population has been divided, with the use of experimental strategies, such as RAPD and multilocus isoenzyme electrophoresis (MLEE), into seven distinct T. cruzi lineages (Marcili et al. 2009; Zingales et al. 2012). The three T. cruzi CL Brener Esmeraldo-like, Dm28c, and marinkelli strains used in this study are from the TcV, TcI, and TcVII lineages, respectively (Marcili et al. 2009; Zingales et al. 2012; Grisard et al. 2014). Thus, the discrepancy observed regarding numbers of members for the Hsp70 and J-proteins is a result of the genetic diversity displayed by the various T. cruzi lineages, highlighting the need for further assessment of these complements in T. cruzi.

Members of each of the J-protein subfamilies (I–IV) were identified in both T. brucei subspecies, though the majority of the J-protein family were found to comprise type III J-protein members. Despite this overwhelming number of J-proteins in T. brucei, very few of these have been biochemically characterized to date. RNAi interference of several of the J-proteins in T. b. brucei were shown to be lethal at one or more stages of the parasite lifecycle, highlighting that the proteins may perform roles unique to the biology of the parasite. Comparative analysis of the T. brucei J-proteins in relation to the selected organisms of this study was conducted to infer cellular function, and potential Hsp70-J-protein partnerships. Obviously, many of the inferences stated in this study will need to be confirmed experimentally. However, it has become increasingly evident that the Hsp70/J-protein machinery is essential to the survival, pathogenicity, and differentiation of the parasite. However, the molecular details of the Hsp70/J-protein chaperone interactions and pathways need to be further elucidated, as some of these pathways may represent a novel means of chemotherapeutic intervention for African trypanosomiasis.

Funding information

This work was funded by a grant from the National Research Foundation (NRF), grant number 87663. S.J.B. is the recipient of an NRF Doctoral Innovation Scholarship. M.J. is the recipient of an NRF DAAD Fellowship.